Nasopharyngeal carcinoma (NPC) is one of the most common cancers in Southeast Asia. Although the frequency of NPC in the American Caucasian population is low, the incidence is high among Southeast Asian individuals who immigrate to the US and reside in Southern California. When NPC recurs or is resistant to initial therapy, the prognosis is grim with no effective treatment available. NPC is closely associated with latent infection by Epstein-Barr virus (EBV). The limited viral gene expression during latency allows EBV to evade immune recognition. However, during the lytic replication phase, the virus fully expresses its genome of approximately 100 genes, a process that leads to cell death in addition to the production of new virions. Therefore, reactivation of the latent virus is a promising strategy to kill the infected tumor cell and increase the host's ability to mount immune responses. Our data show the proteasome inhibitor bortezomib can reactivate EBV from latently infected cell lines with high efficiency. Neo-expression of the lytic thymidine kinase and phosphotransferase genes can then allow tumor-specific phosphorylation/activation of gancyclovir with inhibition of viral and cellular DNA replication and cell death. Thus, re-activation of lytic replication by bortezomib has the potential for killing tumor cells by several mechanisms: 1) Bortezomib-induced apoptosis; 2) death due to viral replication; 3) Induction of anti-EBV immune responses; 4) effects on DNA by concurrently administered gancyclovir. However, although theoretically controlled with gancyclovir, re-activation of viral replication in immunosuppressed patients with cancer could lead to high grade viremia. Thus, this phase I/feasibility clinical trial will test the safety of combining bortezomib and gancyclovir in patients with recurrent NPC, specifically asking whether infection can be controlled and if lytic reactivation of EBV can be identified. If successful, it would form the basis for larger phase II studies to more accurately examine anti-tumor efficacy.